Pivotal Response Intervention Model

Description

Information on Pivotal Response Training for children with Autism.

The Pivotal Response Intervention Model for children with Autism was developed primarily by Drs. Bob and Lynn Koegel at the University of California at Santa Barbara (UCSB). Dr. Bob Koegel trained under Dr. Ivar Lovaas at UCLA, where he received his Ph.D. in Developmental and Clinical psychology. Dr. Lynn Kern Koegel trained in psychology and as a speech and language pathologist (CCC-SLP).

Their early efforts to sort out the characteristics of effective intervention techniques for autistic children is summarized in their excellent book, Teaching Children with Autism: Strategies for Initiating Positive Interactions and Improving Learning Opportunities. Based on these experimentally documented effective intervention techniques, the Pivotal Response intervention model (also known as “Pivotal Response Training” or PRT) was developed. The underlying observation which led to the development of the Pivotal Response intervention model is that there are some skills which are "pivotal” for an individual to function. Addressing a pivotal area of deficit results in gains by the child in areas that were not directly targeted by the initial intervention. The Koegel’s define pivotal areas as “areas that are central to wide areas of functioning such that improvements will occur across a large number of behaviors.” Intervening in a pivotal area produces “large collateral improvements in other areas” (1999a).

Based on their observation of improvement in these collateral areas, they have defined several pivotal areas as potential focal points for intervention including responsivity to multiple cues, motivation to initiate and respond appropriately to social and environmental stimuli and self-direction of behavior including self-management and self-initiations of behavior. A full review of these pivotal areas can be found in the first of a two-part article published by the Koegel’s in 1999 (1999a). Two pivotal areas will be addressed further in this review: motivation and child initiations.

Child Motivation

As used by the Koegel’s, “motivation” is assessed by “observable characteristics of a child’s responding”. An improvement in motivation is defined as “an increase in responsiveness to social and environmental stimuli”. Characteristics which indicate higher motivation include increases in the number of responses a child makes to teaching stimuli, decreases in response latency and changes in affect such as interest, enthusiasm or happiness. An early study by R. L. Koegel, O’Dell and Koegel (1987) showed that interventions designed to improve the rate and accuracy of responding (i.e., child choice, frequent task variation, reinforcing the child’s attempts to comply and incorporating turn-taking,) were also effective in producing generalized and spontaneous verbal language acquisitions in nonverbal children, both in the clinic setting and in the natural environment. In contrast, when language intervention sessions were conducted without the motivational variables, gains in language were considerably reduced. This observation led to inclusion of proven motivational techniques as the central core of all pivotal response interventions. The key motivational variables utilized in Pivotal Response intervention are child-choice, use of natural reinforcers and reinforcement of attempts.

Child-Choice: “Child choice” refers to designing interventions around materials or topics for which the child expresses a preference. This can be accomplished by allowing the child to select stimulus materials from a pool of potential candidates or by selecting a known preferred object to teach a desired skill (i.e., using candy wrapped in colored wrappers to teach colors rather than color flash cards). Efforts are made to incorporate child-choice into routine activities (i.e., selecting which shirt to wear or which food to eat).

Natural Reinforcers: In the Pivotal Response intervention model, the planned reward for the child is something that flows naturally from the child’s actions or verbalization. This contrasts sharply with arbitrary reinforcers (i.e., M&Ms), which are used extensively in the Lovaas-type Discrete Trial Training format. Koegel and Williams (1980) found that children rapidly acquired tasks only if the reinforcer was directly related to the task in a logical way. This principal was demonstrated in an experiment that consisted of teaching the skill of opening a clear glass bottle. If there was candy in the bottle, the child rapidly acquired the skills needed to open the bottle. If the child was asked to open the lid of an empty container with the contingent reinforcer being a food reward given at the successful completion of the task, the child did not comply with the request. The children showed rapid acquisition only when the target behavior was a direct part of the chain leading to the reinforcer. Translated to the naturalistic environment, a child could be asked to say “Root Beer” in order to obtain a soda, or “out” to get out of the car upon arriving at a destination. The reinforcers (Root Beer or getting out of the car) are both direct consequences of the child’s verbalization.

Reinforcement of Attempts: Early models of applied behavior analysis introduced as intervention for autistic individuals required the child to perform the desired activity in a closely proscribed manner. Rewards were not given if the child failed to produce the exact behavior required. PRT provides reinforcement whenever the child makes any unambiguous attempt to produce the desired behavior (referred to as “loose shaping criteria). Less restrictive reinforcement criteria may decrease the “learned helplessness” response that some researchers believe cause autistic individuals to stop trying to respond to requests or prompts.

Child Initiations

The language characteristics of autistic children often include low-levels or complete absence of question-asking, apparent low levels of curiosity and the use of language to obtain desired objects but not to initiate conversation. Typically developing children, in contrast, use questions frequently to initiate social interaction and provide opportunities for language enrichment. Lynn Kern Koegel (1998) demonstrated that autistic children could be taught to ask a simple question (“What’s that?”). The children were able to generalize the question to other appropriate circumstances. Similarly, the question, “What happened?” was associated with increased ability to learn verbs and increased mean length of utterances. Other techniques to increase self-initiations include utilizing additional “Wh” questions and learning to ask for help. An interesting finding was that naïve observers do not judge children as being pragmatically appropriate if they do not show initiations during unstructured interactions. This suggests that question asking is crucial for social interactions.

Research Validation of PRT

A criticism leveled at many of the available intervention programs for autistic children is the lack of randomized trials with adequate longitudinal follow-up data. Although each of the components of the Pivotal Response intervention model has been extensively tested, there are no randomized trials comparing PRT to any other intervention model. The only published follow-up study of PRT was done retrospectively. The study population for the Phase I study consisted of 6 children who had received 4 years of roughly similar intervention consisting of speech therapy (20 minutes twice weekly), one-on-one intervention from graduate students associated with the UCSB Autism clinic in the clinic setting and a community/home based program developed through the autism clinic in conjunction with the child’s classroom teacher, other school and community service providers and parents. The 6 children were chosen to represent “exceptionally favorable intervention outcomes” (3) and “exceptionally poor” outcomes. The children were of similar age at entry to the intervention program (3.1-3.10 years) with similar language levels (2.0-2.6 years). Archival video data was analyzed, with the specific purpose of determining if the children differed with respect to child initiations at the time of entry into intervention. The Koegel’s found that the number of spontaneous initiations at entry was highly correlated with favorable outcome. They then selected 4 children with similar characteristics to the poor outcome children in the Phase I study for prospective intervention with respect to spontaneous initiations. Through a program centered on teaching the children to ask questions (i.e., “What’s that?”), the number of child initiations was dramatically increased for all 4 of the children. Follow-up was conducted for an average of 6 years. At follow-up, 3 of the 4 children were in regular education classes; none of the 4 children in the intervention group continued to qualify for a diagnosis of autism. Unfortunately, there was no control group. Current research at the UCSB autism clinic (supported by a 5 year NIMH grant) is centered on developing parent centered education programs.

Resources:

Educating Children with Autism: Published by the National Academy Press (2001). Edited by Dr. Catherine Lord. Contains the current state-of-the-art summary of research on intervention programs for children with autism, ages 0-8. Available on-line or can be ordered from the National Academy Press website.

Special edition of The Journal of the Association for Persons with Severe Handicaps (JASH) focusing on interventions for young children with autism (volume 24, number 3, Fall 1999): Includes the 1999 two-part article by the Koegel’s on PRT. Can be ordered through the Special Needs Project for $10 (1-800-333-6867, Fax 805-962-5087)

Lisa Benaron, MD completed a combined residency in Internal Medicine and Pediatrics at Chapel Hill, NC. Her interest in autism started when she became a step-mother to a 3 year-old boy (Jake) with classic autism. She is currently the Medical Director at Far Northern Regional Center in Northern California (one of 23 centers devoted to providing services for individuals with developmental disabilities).

References

Koegel, R.L., O’Dell, M.C., and Koegel, L.K. (1987). A natural language teaching paradigm for nonverbal autistic children. Journal of Autism and Developmental Disabilities, 17: 187-200. [151]
Koegel, L.K., and R.L. Koegel (1999a). Pivotal response intervention I: Overview of approach. Journal of the Association for the Severely handicapped, 24:174-185. [148]
Koegel, L.K., and R. L. Koegel (1999b). Pivotal response intervention II: Preliminary long-term outcome data. Journal of the Association for Persons with Severe Handicaps, 24:186-198. [149]
Koegel, L.K., Camarata, S.M., Valdez-Menchaca, M.C., and Koegel, R.L. (1998). Setting generalization of question-asking by children with autism. American Journal on Mental Retardation, 102: 346-357. [147]
Koegel, R.L., and Koegel, L.K. (Eds.). (1995). Teaching Children with autism: Strategies for initiating positive interactions and improving learning opportunities. Baltimore: Paul H. Brookes [150]